1 
Introduction
Bioelectrical Signal Processing
The Human Body
The Biomedical Signal: Reflections of a Secret
Origin of Bioelectrical Signals
Why this Textbook?
Where is Bioelectricity Measured?
The Brain, Heart and Muscles and their Electrical Activity
Electrical Signals  Spontaneous
Electrical Signals  Stimulation
Multimodal Signal Recording

2 
Purposes of Biomedical Signal Processing
Signal Processing Constrains
Noisy Signal Situation  Stress Testing

3 
The Biomedical Signal Processing Challenge
Why Simulated Signals?
The Brain
Neurons = Nerve Cells
Inside the Neuron
Cerebral Cortex and its Lobes
The Cerebral Cortex: Some Basic Facts
The Nervous System
Electroencephalogram  EEG
EEG  (Un)synchronized Activity
EEG acquisition
Important EEG Rhythms
EEG Signals  Examples
Alpha, Beta and Blink Artifacts
The Use of EEG Today
Onset of Epileptic Seizure

4 
Brain Computer Interface
BCI Principle
Brain Computer Interface
BrainGate
EEG Modeling Aspects
Noise and Artifacts
EMG in the EEG
Eye Movements and the EEG
"Optimal" Noise Rejection
"Eye Movement" Electrodes
Noise Rejection by Weighting

5 
Linear Weighting
Weights and the MSE
MSE Minimization
Assumption of Stationarity
Noise Rejection by Weighting
Assumption of Stationarity
Correction of ElectroOculoGram
Adaptive Noise Rejection
MSE Criterion Minimization
Noise Rejection  Filtered Reference Signals

6 
Spectral Analysis of the EEG
FourierBased Spectral Analysis
The Periodogram
Properties of the Periodogram
Periodogram and Segmentation
Spectral Parameters
Trending of Spectral Parameters
EEG and AR Modeling

7

Autoregressive (AR) Modeling
AR Modeling and Linear Prediction
AR Parameter Estimation
The Normal Equation
Multivariate AR Models
EEG Activity Propagation after Finger Movement
AR Modeling and Sampling Rate

8

AR Modeling and Sampling Rate
Adaptive EEG Segmentation
Segmentation  Sliding Window
Criterion for EEG Segmentation
Spectra of a Segmented Signal
Spectral Analysis of Nonstationary Signals
Criterion for EEG Segmentation
Spectral Analysis of Nonstationary Signals
ShortTime Fourier Transformation
Photic Stimulation at Different Rates
Heart Surgery of an Infant

9

STFT and Beyond
TimeVarying AR Model
TimeVarying AR during Seizure
WWDbased TimeFrequency Analysis
The Ambiquity Function
Ambiquity Function, cont't
The Analytical Signal
Analytical Signal in Math Terms

10

Ambiguity Function, cont'
WignerVille Distribution (WVD)
Comparison of STFT and WVD
The Pseudo WVD
CrossTerm Reduction
CWD and 2Component Signal

11

Evoked Potentials
EP = A Transient Waveform
Examples of Evoked Potentials
EP  Definitions
Auditive Evoked Potentials  AEPs
Visual Evoked Potentials  VEPs
Somatosensory Evoked Potentials  SEPs
SEPs during Spinal Surgery
EP Scalp Distribution
Brainstem Auditive EP in Newborns
BAEPs of Healthy Children

12

Cognitive EPs
Ensemble Formation
Formation of an EP Ensemble
10 Superimposed EPs
Model for Ensemble Averaging
Noise Assumptions

13

Ensemble Averaging
Noise Variance
Noise Assumptions
Reduction of Noise Level
Exponential Averaging

14

Introduction
Noise Reduction of EPs with Varying Noise Level
Weighted Averaging

15

Weighted Averaging, cont'
Weighted Averaging: An Example
Robust Waveform Averaging
The Effect of Latency Variations
Lowpass Filtering of the Signal
Latency Variation and Lowpass Filtering
Techniques for Correction of Latency Variations
Estimation of Latency
Woody's Method
Woody's Method: Different SNRs
Noise Reduction by Filtering
Wiener Eiltering
Filtering of Evoked Potentials
Limitations of Wiener Filtering

16

Problem Solving

17

Problem Solving

18

Tracking of EP Morphology
Selection of Basis Functions
Orthogonal Expansions
Basis Functions: An Example
Calculation of the Weights
MeanSquare Weight Estimation

19

Truncated Expansion
Examples of Basis Functions
Sine/Cosine Modeling
MSE Basis Functions
KarhunenLoeve Basis Functions
KL Performance Index

20

KarhunenLoeve Basis Functions
How to get Rx?
Example: KL Basis Functions
TimeVarying Filter Interpretation
Modeling with Damped Sinusoids
Adaptive Estimation of Weights
Estimation Using Sine/Cosine
Estimation Using KL Functions
Limitations

21

Wavelet Analysis
Wavelet Applications
The Correlation Operation
the Mother Wavelet
The Wavelet Transform
The Scalogram
The Discrete Wavelet Transform
Multiresolution Analysis

22

Multiresolution Analysis, cont'
Multiresolution Analysis Exemplified
The Scaling Function
The Approximation Signal x0(t)
Multiresolution Analysis
The Approximation Signal xj(t)
The Multiresolution Property

23

The Wavelet Function
The Wavelet Series Expansion
Multiresolution Signal Analysis: A Classical Example
The Refinement Equation
The Haar Scaling Function
Maar Multiresolution Analysis
Haar Scaling and Wavelet Functions
Computational Coefficients
Filter Bank Implementation
DWT Calculation
Inverse DWT Calculation
Coifflet Multiresolution Analysis
Scaling Coefficient in Noise
The Wavelet Series Expansion
Denoising of Evoked Potentials
EP Wavelet Analysis

24

The Electrical Activity of the Heart
The Heart ...
Blood Flow od the Heart
Conduction System of the Heart
Cardiac Excitation
Electrical Vectors of the Heart
Cardiac Excitation
The Cardiac Cycle & Wave Shape
Extremity Leads  I, II, III
Årecordial Leads  V1 to V6
The Standard 12Lead ECG

25

ECG Waves: PQRST
Normal Sinus Rhythms
Heart Rate Variability
Arrhythmias: Ectopic Beats
Arrhythmias: Bi & Trigenimy
Arrhythmias: Atrial Flutter/Fibrillation
Arrhythmias: Ventricular Flutter/Fibrillation

26

Heart Attack
Myocardial Ischemia
Noise in the ECG
Clinical ECG Applications
The Exercise Stress Test
Stress Testing and Ischemia  ECG Reaction 1
Stress Testing and Ischemia  ECG Reaction 2
ST Reaction Versus Heart Rate  Decision Regions
HighResolution ECG and Cardiac Late Potentials
Spectral Analysis of the ECG?
Spectral Analysis of the Heart Rate?
EEG, EP, and ECG: Time Base?
ECG Signal Processing
ECG Filtering Techniques ...
ECG Baseline Wander

27

Discussion 
28

Baseline Filtering: Phase Aspects
Baseline Filtering: An Example
Cubic Spline Interpolation
50/60 Hz LTI Notch Filter
Nonlinear 50Hz Filtering
Nonlinear Filtering Exemplified
50/60 Hz Filtering

29

50/60 Hz Filtering
QRS Detection Problems
Spectral Detection Problems
QRS Detection
Models for QRS Detection
Design of Linear Detection Filter
Simple Detector Filter Structures
Simple Filters for QRS Detection  Frequency Response
Design of Nonlinear Transformation
Envelopebased Detection
Envelope Examples
Preprocessor Output
QRS Detection: Decision Rule
QRS Detector Performance

30

QRS End Delineation
LPDbased Delineation
Data Compression
ECG Data Redundancy
Data Compression of ECG Signals
Lossless Data Compression based on Linear Prediction
Linear Prediction
Lossy Data Compression
Example of AZTEC
The SAPA Principle
Example of SAPA
KLTbased Data Compression
KLT Compression with Tolerance
KLT using Universal Data
KLT using SubjectSpecific Data
Handling Interbeat Redundancy
Performance Measures

31

What's behind the Beat?
Heart Rate Variability (HRV)
Conductor's Heart Rate
Heart Rate and Respiration
RR Interval Series
24hour RR Interval
24h RR Interval Histograms
The RR Interval Series  the Tachogram
Heart Rhythm Representations
Lowpass Filtered Event Series

32

Integral Pulse Frequency Modulation (IPFM) Model
Output of the IPFM Model
Heart Timing Signal
Heart Timing Signal: Example
Why Spectral Analysis of HRV?
Spectrum of Counts
Lomb's Periodogram
Lomb's Periodogram and the Classical Periodogram
HRV Spectral Analysis
HRV Spectrum: A Comparison
Power Spectrum with One Modulation Frequency (0.16 Hz)
Power Spectrum with Two Modulation Frequencies
HRV and Sudden Cardiac Death

33

Ectopic Beat Correction
Do not maltreat the signals ...
